Thermal Storage - Keeping cool after dark

Just a tiny fraction of work in the office sector is new build. The vast majority is refurbishment, so real inroads in cutting CO₂ emissions will only be made by upgrading the efficiency of existing buildings.

However, refurbishing buildings with mechanical cooling can be difficult and costly in terms of money and environmental impact. This has lead to the increasing use of passive alternatives such as night cooling. An innovative new technology, in the form of Airdeck, promises gold stars in the shape of energy savings and straightforward retrofitting.

Night cooling relies on the circulation of outside air through a building at night. The cold air comes into contact with the building fabric, thereby drawing out and releasing its heat energy. This helps to offset heat gains and limit temperature rises the following day.

Floor slabs are normally used as thermally massive stores, typically coming into contact with the ventilation air through the exposed underside of the slabs. A possible alternative is to supply air through the floor void to bring it into contact with the top of the slab.

In both cases the slab surface needs to be exposed, something that is often difficult when refurbishing an existing building. Exposing the slab surface can also be aesthetically and acoustically undesirable, and occupants have no control over heat flow into and out of the slab. This can result in increased winter heating demand, with the slab acting as a store for heat losses at night.

What's more, the rate of heat transfer between the air and the slab surface is low compared with the slab's full storage capability, which limits the performance of the exposed surface techniques. The real challenge has been to devise a solution for transferring heat efficiently between air and slab without having to expose the slab surface.

The Airdeck concept

The Airdeck concept is to bring air into close contact with the slab in order to improve surface heat transfer. Fixing Airdeck to the slab surface forms narrow paths for air to pass through. Airflow within these paths tends to be highly turbulent, providing good heat transfer performance between the air and slab.

Heat flows are determined not only by heat transfer at the surface, but also by the slab's conductive storage capability. Normally the surface heat transfer is much lower, which limits the heat flow.

This project aimed to increase the surface heat transfer above the conductive storage capability of a slab (typically 15 W/m²K) while not doubling the value. This balances the improvement in heat transfer (which at higher values will be limited by the conductive storage capability of the slab) against the increased fan pressure required.

The airpath height is a key parameter determining surface heat transfer. This was the focus of the first stage of the development work.

Testing the kit

A test rig was built at Oxford Brookes University to measure the heat transfer coefficients and associated pressure drops for a range of standard sheet profiles. Findings were used to establish the parameters for a full-sized module which was built and tested at the Cranfield Institute of Technology.

In figure 1 results from the Airdeck module tests are shown alongside curves of theoretical values for smooth and rough tubes. The rough tube values assume worst case average surface roughness of 1 mm in the airpaths.

CIBSE Guide C41 equations C4·5 and 4·8 have been used to calculate the friction factor, taking into account the surface roughness. Equation C4·4 was used to calculate the pressure drop per unit length (figure 2). Heat transfer coefficients for the smooth tube case have been calculated from CIBSE Guide C32 equation C3·21:

Nu = 0·023 x Re^0·8 x Pr^0·33

where: Nu is the Nusselt number, Re is the Reynolds number and Pr is the Prandtl number.

The Reynolds-Colburn analogy³ was used to calculate the heat transfer coefficient for a rough tube:

St.Pr^2/3 = f/2

where:

St is the Stanton number hc/(r.u.cp);

Pr is the Prandtl number;

f is the friction coefficient;

hc is the convective heat transfer coefficient (W/m²K);

r is the air density (kg/m³);

u is the air flow velocity (m/s);

cp is the air specific heat capacity (J/kgK).

As can be seen in Figure 1, the test results lie between the two theoretical extremes. A worst case design approach – assuming smooth tube values for the heat transfer coefficient and rough tube values for the pressure drop – was used to cope with uncertainty in the surface roughness.

Two other key design parameters are the length of the airpath and the area (or width) of Airdeck required. Suitable values have been determined by modelling a typical application – assuming an air circulation rate of 6 ac/h.

The length of the airpath determines how efficiently the system exchanges heat between air and slab. The longer the airpath, the closer the air exit temperature will be to the slab temperature (ie the nearer it will be to 100% heat exchange efficiency). However, diminishing returns mean that efficiency improvements from longer airpaths need to be balanced against cost.

The area of Airdeck required is a function of the amount of heat energy which can be stored in the slab. This is limited by the volume of air brought into the building. To minimise costs, the Airdeck area should be no greater than what is needed to dissipate the energy stored in the building fabric.

Night cooling is most appropriate for buildings with moderate daytime heat gains, such as offices. The Airdeck solution is suitable for both retrofit and new installation into floor and ceiling voids. The prototype was designed to co-ordinate with floor pedestals and ceiling hanger systems located on a 600 mm grid.

Installation

For both floor and ceiling void applications the mechanical ventilation system may be centralised or set-up around a local zone system. In either case, low system pressure drops are desirable to minimise fan energy consumption and fan pick-up (heating of the airstream by the supply fan). Air flow rates, at 3-6 ac/h, are similar to those for typical mechanical ventilation systems.

Once installed, supply air would be ducted to the Airdeck from an air handling unit or supply fan. For floor void installations the profiled sheets would normally have a free air outlet – the floor void being used as a supply plenum. Supply air would pass from the floor void into the conditioned space via floor diffusers. For ceiling void installations, outlet from the profiled sheets would normally be ducted to ceiling supply diffusers.

Airdeck shows enormous potential as a refurbishment solution in false ceilings. As well as the Cranfield-tested steel sheet design, a prototype has been produced using plasterboard. This relies on smaller, lighter modules for installation within a ceiling void.

Airdeck's performance looks good against other night cooling techniques (table 1). When installed behind a false ceiling, FACET modelling shows it to outperform an exposed slab and a reference case with no thermal link between air and slab. In the real world, these figures should translate into more comfortable, more productive offices.